One priority. Key results you can score on the last day of the quarter.
A quarter's objectives and key results drafted from one priority and your current numbers, with every task rewritten as a result that moves.
Most OKR sets are project lists wearing a different label. Launch the page, hire the person, run the campaign. All can be completed while nothing improves, so the end-of-quarter review becomes an argument about what counted.
OKR Draft starts from a single priority and derives no more than three objectives. Every key result must have a metric, a baseline, a target, and a date. It rejects tasks, marks what is committed and what is stretch, and assigns an owner and a weekly question.
What it produces.
- →One to three objectives tied to your single priority
- →Measurable key results with baselines, targets, and dates
- →A separate task list for work that did not qualify as a result
- →Owners and a weekly check-in question for each key result
- ✦Founders setting the company's goals for the quarter
- ✦Team leads who need goals that can be scored
- ✦Anyone whose last quarter's OKRs were never checked
- →okr-draft.zip: a native Claude Skill. Upload to Settings → Skills in Claude, or unzip to
~/.claude/skills/ - →okr-draft-skill.md: plain markdown. Works in ChatGPT Custom GPTs, Gemini, Claude Projects, agent system prompts, anywhere
- →A README inside the zip with install steps for Claude, ChatGPT, and Gemini
What it does with a real question.
I'm the head of customer success at a 40-person B2B scheduling software company, and we need Q3 OKRs by July 8. Our one priority is cutting churn: logo churn hit 4.1% last quarter, up from 2.8%, mostly in accounts under $5K ARR. Last quarter's goals were basically a task list and nobody scored them at the end.
Start with the number you already have. Your priority is one sentence: cut logo churn among accounts under $5K ARR. I would build two objectives from it, something like "Keep small accounts from drifting away unnoticed" and "Fix whatever is making them leave," with two or three key results each. Each needs a baseline, and you have one: 4.1% last quarter, 2.8% the quarter before. A scoreable key result would read "Cut quarterly logo churn from 4.1% to 3.2% by September 30," with a stretch version at 2.8%, a full return to the earlier rate. Mark 3.2% as committed and 2.8% as stretch, and write down what has to go right for the stretch to land.
Here is where last quarter's list probably went wrong, so apply the task test now: could this key result be completed without churn improving? "Launch a health score" can be. "Reach every sub-$5K account with no admin login in 14 days within three business days, at 90% coverage" cannot be faked, because it measures a result you can pull from data. Treat the health score as a project that supports that key result, and keep it on a separate list.
Before July 8, split last quarter's churned logos by cancellation reason and tenure, even roughly. If most left in their first 90 days, your objectives should lean toward onboarding. If they left at renewal, lean toward early renewal outreach. That one spreadsheet decides which key results deserve a place.
The full Playbook runs the whole procedure and writes the complete deliverable, not this excerpt.
Written by claude-sonnet-5-5 with this Playbook loaded, 2026-10-11. Checked automatically for length and for copied framework text.
No subscription · free updates within v1.x
Works as a Claude Skill (.zip) and as a plain .md for ChatGPT, Gemini & every LLM
Delivered to your email within 60 seconds · Secured by Stripe
Price includes UK VAT where applicable
Before you buy.
How is this different from just asking Claude directly?
Asked cold, Claude writes plausible OKRs full of verbs like launch and improve. This starts from one priority, requires a baseline for every number, and rejects any key result that could be finished without anything getting better.
How is this different from a Tool?
Playbooks are systematically produced to a fixed template and priced accordingly ($12.99). Tools are hand-built, one at a time, and go deeper ($39 to $199). A Playbook gives you a reliable method for routine work; a Tool is for when the judgment matters more than the format.
Will this work with ChatGPT, Gemini, and other LLMs?
Yes. You get a native Claude Skill (.zip) that auto-invokes in Claude, plus a plain .md you can paste into a ChatGPT Custom GPT, a Gemini system prompt, or any LLM. The procedure travels with the file.
Can I share this with my team?
It's licensed for your own personal and professional use. The output it produces is yours to circulate. To put the Playbook itself in the hands of a whole team, email hello@authority.md about team licensing.
What if it doesn't work for me?
If the file hasn't been delivered, or arrives defective or empty, we refund in full: see our refund policy.